Scope
This joint workshop aims on gatehring researchers active in academia and industry to share innovative ideas and discuss future challenges in reproducibility of robotics experiments.
The event will be held virtually spanning in two days: Part I on May 31th and Part II on June 4th, during ICRA2021 .
We will put together the like-minded communities working on reproducibility and benchmarking and in the AI Driving Olympics.
Ideally, participants will help establish realistic mechanisms for the spread and improvement of reproducible and measurable robotics research, with a special focus on issues related to generalization and replication of results.
Moreover, we hope to better understand how the need for reproducibility affects publication formats, problem definitions, and evaluation tools. The IEEE Robotics and Automation solicits R(eproducible)-articles since 2017. So far three such articles have been accepted and their results wait to be reproduced.
You can find useful info about the R(eproducible)-articles here:
R-articles
* So far three R-articles have been published:
- G. Golluccio, G. Gillini, A. Marino and G. Antonelli, "Robot Dynamics Identification: A Reproducible Comparison With Experiments on the KINOVA Jaco2," in IEEE Robotics & Automation Magazine
- F. Bonsignorio and E. Zereik (2020), “A Simple Visual-Servoing Task on a Low-Accuracy, Low-Cost Arm: An Experimental Comparison Between Belief Space Planning and Proportional-Integral-Derivative Controllers”, IEEE Robotics & Automation Magazine.
- S. Carpin (USA) about: J. L. Susa Rincon and S. Carpin (2019), "Time-Constrained Exploration Using Toposemantic Spatial Models: A Reproducible Approach to Measurable Robotics," in IEEE Robotics & Automation Magazine
You can find useful info on the AI Driving Olympics here: AI-Driving-olympics and here: Driving Olympics